EFFECTS OF FIBER ON FRACTURE PROPERTIES OF LIGHT WEIGHT CONCRETE MADE FLY-ASH PELLETIZED AGGREGATES

Transcription

1 Fracture Mechanics of Concrete Structures, Proceedings FRAMCOS-2, edited by Folker H. Wittmann, AEDIFICATIO Publishers, D Freiburg (1995) EFFECTS OF FIBER ON FRACTURE PROPERTIES OF LIGHT WEIGHT CONCRETE MADE FLY-ASH PELLETIZED AGGREGATES Chang, C.Y. and C.L. Hwang, Department of Construction Engineering, National Taiwan Institute of Technology, Taiwan, R.O.C.. M. Shieh, Taiwan Provincial Hsiu-Shui Vocational School, Hsiu-Shui, Chang-Hua, Taiwan, R.O.C. Abstract study presents the experimental results of fracture properties of lightweight concrete made with cold-pelletized aggregates incorporating three steel concentrations, 1 %, 2 % and 3 % of cement weight respectively. Experimental results showed that the increase of fiber amount increased compressive, splitting tensile and strengths, toughness of concrete. Compressive strengths of 100x200 mm cylindrical concrete specimen at 28 days ranged from fiber) to MPa (3% fiber). means of size effect law, it was that the fracture energy increased from Gf = 37.2 Nim (no fiber) to 62.7 Nim (3% fiber), respectively. Predictions of structural response on the three-point bending specimen from R-curves were also presented. 1 Introduction Lightweight concrete is defined as having an oven-dry density range of approximately kglm ]. The compressive strength of lightweight concrete larger than MPa is considered to be the structural concrete by RILEMICEB. this study, fly-ash was used as 803

2 aggregate method [Bijen 1986] ecological advantage of consumption. various fiber reinforcements incorporated concrete mixture to increase the ductility of the concrete [Swamy 1989]. The main function of is to propagation of cracks through the brittle cementitious matrix. is especially important for lightweight concretes, since the strength of aggregate is usually weaker than that cement paste such that aggregates can not offer a crack arrest mechanism, resulting a fracture. An understanding of effects of steel reinforcement on fracture properties of lightweight concrete appears to be study presents experimental results of lightweight aggregate and concrete aggregates. Three different amounts of steel reinforcement, 1 %, 2 % and 3 % of cement weight were used. properties were on the base of size-effect law 2 response by size-effect Test method based on size-effect law is a simple and effective way calculating fracture parameters of concrete. With this method, only measured values experiment are the maximum loads of several sufficiently sizes of specimens with geometrically similar shape. The minimum size range for specimen is 1 to 4 [RILEM 1990]. fracture energy Gf calculated by size-effect law for the Single-Edge Notch-Three-Point-Bending (SENTPB) beam specimens, as shown 1, can be expressed as B2(/, )2 GI = 2 u d<fl(<t.o) cne where E = Young's modulus; en = (1.5 )/[d(l-a) 2 ]; d = span; a = c/d; c = effective crack extension; a 0 = a 0 /d = length of notch; = tensile strength; do and B = unknowns to be g(a) = shape as given in Eq. (2). (1) g(a) = [6.647{a,(l-2.5a +4.49a a a 4 )] 2 (1-a)l.5 (2) The values of B and d 0 are calculated from the maximum failure load Pu' tensile strength fu and nominal stress an through the linear regression 804

3 p 1 1 SENTPB specimen Fig. 2 Fly-ash aggregate Y = AX+C as expressed Eqs. (3) X = d, y = fu 2 a = c Pu. B = _1_. d = C <JN ' N n bd ' [C ' 0 A where b = thickness of specimen. fracture properties determined by (1) to (4) are those values extrapolated from the laboratory specimen to the specimens with sufficiently large sizes. Therefore, G f is assumed to be the material property. On the other it has been vnr-"''1"'1 that the crack resistance (R curve) varies with amount of crack For concrete, as a crack extends a damaged zone, a microcracking zone (fracture process zone) is created ahead the tip initial crack or notch such that crack resistance increases with the crack extension until the peak load reaches, e.g., a completed damaged zone has been developed. Thus crack resistance for concrete specimen is strongly dependent on its geometry. The R-curve equation for concrete from size-effect law can be expressed by [Gettu 1990]: (5) 805

4 equation together with LEFM relations can be used to predict the load-deflection curves for SENTPB beam specimens. uc = deflection at mid-span due to fracture, ub and us = elastic deflections due to bending and shear respectively, and u = uc + ub + u 3 = total displacement of beam specimen at mid-span. The equations required for this calculation are summarized as follows [Gettu 1990]: 2Pf" Uc = - g(t)dt ; Ebo P = b Ed R(c) g(a) (6) pz3 Pl us = 0.6(1 +v )- bde (7) where v = Poisson's Choosing different values of a or c, one can calculate the complete load-deflection curve of the SENTPB beam specimen from (5) to (7). Note that the R-curve value (5) is constant after load, since the fracture process zone is developed a traction-free crack in the post-peak 3 Coarse aggregate The coarse fly-ash lightweight aggregates used in this study were made of the mix compound bottom ash, fly ash (passing through sieve #4), cement (Type I) hydrated lime [Shieh 1994]. Water was the wetting agent acting as such that the wet fly ash would be pelletized with other adhesive through the rolling motion. In this study, the diameters of manufactured fly-ash aggregates ranged from 5 mm to 25 mm. Only passing through sieve #1/2" (12. 7 and retaining at sieve #4 mm) were selected as the coarse aggregates, as shown specific gravity of the aggregate was about 1.33; the average moisture-absorption ratios were about 16.4%, 20.5% and 24.7% at hours and 14 days respectively [Shieh 1994]. Those aggregates were submerged in water for about 30 minutes before into concrete. By placing a piece of pelletized aggregate on a base plate of a material testing machine, and then applying a compressive load gradually from the vertical direction to the particle, the volume average compressive strength of aggregate, a 22, can be calculated by following equation [Chang 1995]: 806

5 where F 2 = vertical load; h = distance between two points applied load. Various sizes of fly-ash aggregates were selected for the compressive test and strength according to (8). The average compressive aggregates was about 12.2 MPa [Shieh 1994]. For comparison purpose, the volume average compressive strength of the normal crushed gravel used high-strength concrete ranged from 94 to 157 MPa using same formula [Chang 1995]. aggregate, admixture and steel fiber Natural river sand shape and highly siliceous composition was The FM of natural sand excluding the size larger than was The specific gravity and moisture absorption ratio of natural sand were 2.68 and 1.75% respectively. The sand met the requirements ASTM C-33. I cement meeting the requirements ASTM C150 was used. dosage of Type F superplasticizer was used as an admixture to properties of concrete. Steel as shown Fig. 3, having a 0.6 a length of 25 mm and...,,..,...,.,... ends which offered strong mechanical anchorage in concrete were used. A small amount those steel fiber reinforcements (SF) with 1 %, 2 % and 3 % of cement weight respectively was mixed concrete. 4 Experimental Basically, the concrete proportioning method used in this study follows similar concepts provided in the specification of ACI standard. The materials of mix proportions for 1 m 3 concrete used in this study were: Cement = 522 kg; water = 157 kg; coarse aggregate = 586 kg; fine aggregate = 636 Type F superplasticizer of 6.53 kg/m 3 was also used for all the Cylindrical concrete specimens with size of cm by 20 cm were used compressive and splitting tensile tests. Beam specimen of 7.5x7.5x22.5 cm was used for the flexural test. concrete specimens were stored in lime water until 24 hours before they were tested. The specimen for compressive strengths were performed by a 2000 KN test machine according to ASTM C469 and C496 respectively. Specimens for strength and fracture property were carried out by a 50 KN MTS machine a close-loop-controlled stroke system. Four different sizes SENTPB specimens ( 4x4xll, 4x8x2 l, 4xl 6x43 and 4x32x85 cm) were used the fracture experiment, tested with loading at a constant strain rate of 0.1 mm/min. Typical is shown Fig. 807

6 3 Short steel fiber Fig. 4 Typical test set-up 5 results of lightweight concrete are shown in Table 1. The 28-day compressive strength ranged from 33.9 MPa (0% SF) to 37.7 MPa (3% of improvement was about 11 %. Increasing the steel fiber content increased the density of concrete and compressive strength, but decreased Poisson's ratio and workability. Average value of Poisson's for high-strength light-weight concrete had been reported as about 0.2 regardless of concrete strength, test age and curing condition [Slate 1986]. this study, the Poisson's ratios v dropped from 0.22 to 0.16, were smaller than the values of 0.23 to 0.32 for normal-weight concrete [Slate 1986]. The reason for this could be due to the fact that void ratio was high for fly-ash aggregate so that it was prone to a larger deformation in axial direction under compression test. splitting tensile strengths f and flexural strength fr ranged from to 3 MPa and from 2.37 to L.53 MPa respectively. Addition of steel increased both strengths to 22.6% and 6.8 % respectively. The ratios splitting tensile to compressive strength and flexural to compressive for normal-weight concrete ranged from about 0.08 to 0.14 and about 0.11 to 0.23, respectively. For the concrete tested in this study, ratios dropped to 0.09 and 0.07 respectively as shown in Table

7 1 of lightweight concrete f ' EC fsp fr Slump v c SF.. 'Ye 3 kg/m MPa GPa Ma MPa cm 0% % % % Notes: (1) Ee = {3.31(fc') }(1'c /2320) GPa; fc' MPa, density of concrete; (2) All test data above were the average of tested specimens at age of 28 days. concrete and enhances concrete a considerable amount of ductility. to those properties, linear elastic fracture mechanics (LEFM) had found to be inadequate to the fiber reinforced concrete [Rossi 1986]. Various nonlinear fracture mechanics models have been proposed to either predict the load-deflection curve or to determine the fracture energy of fiber reinforced concrete [J enq 1986, Hillerborg 1980]. Size-effect model was used in this study for determining the fracture properties of fly-ash light-weight concrete. The fracture test results are shown Table 2. Table 2 Maximum loads P max on SENTPB beam specimens (N) d P max (0% SF) pmax (1 % SF) p max (2% SF) P max (3% #1 #2 #3 #1 #2 #3 #1 #2 #3 #1 #2 #3 cm ~128 ~ ~ During the experiment, it was observed that the failure planes 48 pieces of beam specimens exhibited same failure mechanism, failure planes usually cutting through the aggregates. regression lines by size-effect model for both plain and fiber-reinforced lightweight concretes were shown in Fig. 5. Figs. 6 and 7 showed the relevant effect curves. By means of Eqs. (1) to (4) and data in Table 2, the fracture energies were found to be 37.2, 37.9, 44.4 and 62.7 Nim lightweight concrete with SF content of 0, 1, 2 and 3 % respectively. coefficients of errors (vertical deviations from regression line) were wy1x = 15.3, 16.3, 11.9 and 9.3% respectively. The increase of Gf was 809

8 Y=AX+C 1.8 SF A c 0% % % Q.323 D 1.4 3% N z 1.0 J ':::J % 3% o.o ,-,~~~~~~~~~-,--ri Fig. 5 Mil!llll"'ITll'll d X % % Strength criterion r--r r--o ,-----,~ ,-j log~ Fig. 6 Size-effect curves (1) 0.6 ~ D 2% Strength criterion '+-':::J 'Oz 0.2 -C) % ~~~~~~~~~ ~3000 ~ 2500 ca % Steel fiber Model Experiment 8=1.515 do= mm ~ ~~~~~~~..----< o_.,_,_,.,-,-,-,---,-,-,-,.--.-r-r ~-r-r-ro-r-i Load-point deflection Fig. 7 curves (2) Fig. 8 Load-displacement curves 810

9 about 68.5% (3% SF). For normal strength concrete value of Gf was about 38.4 Nim, this value was strength concrete [Gettu 1990]. Although the reinforcement only made a negligible variation strength, it improved the toughness significantly. a typical load-point deflection for SENTPB shown in Fig. 8. Young's modulus 16.3 GPa, compliance of the beam specimen, and Poisson's ratio...,,.,..., were used in the structural response calculation. 6 Conclusions 1. Although the size-effect law used current study consider the effects of steel fiber reinforcement it does provide a quantitative information on fracture properties of fly-ash aggregate lightweight concrete from various amounts of added fiber. steel fiber only a negligible enhancement on the ""' 1 "Y'lnri::»c and fracture was observed. amount fiber increased to 2 and 3%, the ratio of energy had risen by 19 and 69% respectively. technique is especially useful to develop structural... A..,,.,...,... lightweight aggregates. Other fracture properties length and size of fracture process zone, and tip opening displacement can also calculated effect law. For a constant value of Gf, the structural response...,......,,..._..._... curve values mainly depends on fracture parameters _ 0, etc. and the geometry and material properties of the...,...,...,... implied in the original derivation. the curve predicted from size law individually. Once the values those necessary defined, the R-curves can be easily used to predict 0 "r 0 "!"',.,,...1:',",,."' structural response on the lightweight concrete different amounts of added steel Acknowledgement financial support from NSC E01 appreciated. 811

10 Estimation coarse aggregate......, to be published ACI Material Clarke, J. L. ( Lightweight Aggregate Blackie...,...,..., & Professional, First Ed., Chapman & Gettu, R., brittleness Dec., Martha, E.K. (1990) Fracture properties concrete. ACI Material Journal, Nov- Hillerborg, A. (1980) Analysis of fracture by means of the fictitious crack particularly for fibre-reinforced concrete, International of Cement Composites, 2( 4 ), Jenq, Y.S. and Shah, S.P.(1986) Application of two parameter fracture model to concrete and fiber reinforced concrete, in Fracture Toughness Fracture Energy of Concrete,(ed. Wittmann), RILEM determining Materials Recommendation (1990) Size-effect method for energy and process zone size of Structures, 23, Rossi, et al. (1986) Comparison between plain concrete toughness steel fibre concrete toughness. Cement Research, 16(3), Shieh, M.M. (1994) Effects of fiber reinforcement on engineering properties lightweight concrete, Master Thesis, National Taiwan Institute of Technology, Taiwan (in Chinese). Slate, F.O., Nilson, and Marthinez, S. (1986) Mechanical properties of high-strength light-weight concrete. ACI Journal, Swamy, R.N. and Barr, (1989) Fiber Reinforced Cements and Concretes: Recent Developments. Elsevier Science Publishers Ltd. 812